x86_64: Fix svml_s_acosf4_core_sse4.S code formatting

This commit contains following formatting changes

1. Instructions proceeded by a tab.
2. Instruction less than 8 characters in length have a tab
   between it and the first operand.
3. Instruction greater than 7 characters in length have a
   space between it and the first operand.
4. Tabs after `#define`d names and their value.
5. 8 space at the beginning of line replaced by tab.
6. Indent comments with code.
7. Remove redundent .text section.
8. 1 space between line content and line comment.
9. Space after all commas.

Reviewed-by: Noah Goldstein <goldstein.w.n@gmail.com>
This commit is contained in:
Sunil K Pandey 2022-03-07 10:47:08 -08:00
parent 5f7fb3ea48
commit c321692af7

View File

@ -29,242 +29,241 @@
/* Offsets for data table __svml_sacos_data_internal
*/
#define SgnBit 0
#define OneHalf 16
#define SmallNorm 32
#define MOne 48
#define Two 64
#define sqrt_coeff 80
#define poly_coeff 112
#define Pi2H 192
#define PiH 208
#define SgnBit 0
#define OneHalf 16
#define SmallNorm 32
#define MOne 48
#define Two 64
#define sqrt_coeff 80
#define poly_coeff 112
#define Pi2H 192
#define PiH 208
#include <sysdep.h>
.text
.section .text.sse4,"ax",@progbits
.section .text.sse4, "ax", @progbits
ENTRY(_ZGVbN4v_acosf_sse4)
subq $72, %rsp
cfi_def_cfa_offset(80)
subq $72, %rsp
cfi_def_cfa_offset(80)
/* X<X^2 iff X<0 */
movaps %xmm0, %xmm14
/* X<X^2 iff X<0 */
movaps %xmm0, %xmm14
/*
* 2*sqrt(X) ~ Sh - Sl (to 24+ bits)
* SQ ~ 2*sqrt(X)
*/
movups __svml_sacos_data_internal(%rip), %xmm3
movups OneHalf+__svml_sacos_data_internal(%rip), %xmm5
/*
* 2*sqrt(X) ~ Sh - Sl (to 24+ bits)
* SQ ~ 2*sqrt(X)
*/
movups __svml_sacos_data_internal(%rip), %xmm3
movups OneHalf+__svml_sacos_data_internal(%rip), %xmm5
/* x = -|arg| */
movaps %xmm3, %xmm4
orps %xmm0, %xmm4
/* x = -|arg| */
movaps %xmm3, %xmm4
orps %xmm0, %xmm4
/* Y = 0.5 + 0.5*(-x) */
movaps %xmm5, %xmm6
mulps %xmm4, %xmm6
/* Y = 0.5 + 0.5*(-x) */
movaps %xmm5, %xmm6
mulps %xmm4, %xmm6
/* x^2 */
movaps %xmm4, %xmm13
mulps %xmm4, %xmm13
addps %xmm6, %xmm5
/* x^2 */
movaps %xmm4, %xmm13
mulps %xmm4, %xmm13
addps %xmm6, %xmm5
/* SQ ~ 2*sqrt(Y) */
rsqrtps %xmm5, %xmm8
minps %xmm5, %xmm13
movaps %xmm5, %xmm2
movaps %xmm13, %xmm1
cmpltps SmallNorm+__svml_sacos_data_internal(%rip), %xmm2
cmpnltps %xmm5, %xmm1
cmpltps %xmm13, %xmm14
addps %xmm5, %xmm5
andnps %xmm8, %xmm2
movaps %xmm13, %xmm11
movaps %xmm2, %xmm9
movaps %xmm1, %xmm6
mulps %xmm2, %xmm9
andnps %xmm4, %xmm6
mulps %xmm5, %xmm2
mulps %xmm13, %xmm11
mulps %xmm9, %xmm5
movups sqrt_coeff+__svml_sacos_data_internal(%rip), %xmm10
andps %xmm0, %xmm3
/* SQ ~ 2*sqrt(Y) */
rsqrtps %xmm5, %xmm8
minps %xmm5, %xmm13
movaps %xmm5, %xmm2
movaps %xmm13, %xmm1
cmpltps SmallNorm+__svml_sacos_data_internal(%rip), %xmm2
cmpnltps %xmm5, %xmm1
cmpltps %xmm13, %xmm14
addps %xmm5, %xmm5
andnps %xmm8, %xmm2
movaps %xmm13, %xmm11
movaps %xmm2, %xmm9
movaps %xmm1, %xmm6
mulps %xmm2, %xmm9
andnps %xmm4, %xmm6
mulps %xmm5, %xmm2
mulps %xmm13, %xmm11
mulps %xmm9, %xmm5
movups sqrt_coeff+__svml_sacos_data_internal(%rip), %xmm10
andps %xmm0, %xmm3
/* polynomial */
movups poly_coeff+__svml_sacos_data_internal(%rip), %xmm12
movaps %xmm1, %xmm15
mulps %xmm13, %xmm12
subps Two+__svml_sacos_data_internal(%rip), %xmm5
mulps %xmm5, %xmm10
addps poly_coeff+16+__svml_sacos_data_internal(%rip), %xmm12
mulps %xmm2, %xmm5
mulps %xmm11, %xmm12
addps sqrt_coeff+16+__svml_sacos_data_internal(%rip), %xmm10
mulps %xmm5, %xmm10
movups poly_coeff+32+__svml_sacos_data_internal(%rip), %xmm5
subps %xmm10, %xmm2
mulps %xmm13, %xmm5
movups MOne+__svml_sacos_data_internal(%rip), %xmm7
andps %xmm1, %xmm2
cmpnleps %xmm4, %xmm7
addps poly_coeff+48+__svml_sacos_data_internal(%rip), %xmm5
movmskps %xmm7, %edx
orps %xmm2, %xmm6
addps %xmm12, %xmm5
mulps %xmm13, %xmm5
pxor %xmm3, %xmm6
movups PiH+__svml_sacos_data_internal(%rip), %xmm7
andps %xmm1, %xmm7
addps poly_coeff+64+__svml_sacos_data_internal(%rip), %xmm5
mulps %xmm13, %xmm5
andps %xmm14, %xmm7
mulps %xmm6, %xmm5
andnps Pi2H+__svml_sacos_data_internal(%rip), %xmm15
addps %xmm5, %xmm6
addps %xmm15, %xmm7
addps %xmm6, %xmm7
testl %edx, %edx
/* polynomial */
movups poly_coeff+__svml_sacos_data_internal(%rip), %xmm12
movaps %xmm1, %xmm15
mulps %xmm13, %xmm12
subps Two+__svml_sacos_data_internal(%rip), %xmm5
mulps %xmm5, %xmm10
addps poly_coeff+16+__svml_sacos_data_internal(%rip), %xmm12
mulps %xmm2, %xmm5
mulps %xmm11, %xmm12
addps sqrt_coeff+16+__svml_sacos_data_internal(%rip), %xmm10
mulps %xmm5, %xmm10
movups poly_coeff+32+__svml_sacos_data_internal(%rip), %xmm5
subps %xmm10, %xmm2
mulps %xmm13, %xmm5
movups MOne+__svml_sacos_data_internal(%rip), %xmm7
andps %xmm1, %xmm2
cmpnleps %xmm4, %xmm7
addps poly_coeff+48+__svml_sacos_data_internal(%rip), %xmm5
movmskps %xmm7, %edx
orps %xmm2, %xmm6
addps %xmm12, %xmm5
mulps %xmm13, %xmm5
pxor %xmm3, %xmm6
movups PiH+__svml_sacos_data_internal(%rip), %xmm7
andps %xmm1, %xmm7
addps poly_coeff+64+__svml_sacos_data_internal(%rip), %xmm5
mulps %xmm13, %xmm5
andps %xmm14, %xmm7
mulps %xmm6, %xmm5
andnps Pi2H+__svml_sacos_data_internal(%rip), %xmm15
addps %xmm5, %xmm6
addps %xmm15, %xmm7
addps %xmm6, %xmm7
testl %edx, %edx
/* Go to special inputs processing branch */
jne L(SPECIAL_VALUES_BRANCH)
# LOE rbx rbp r12 r13 r14 r15 edx xmm0 xmm7
/* Go to special inputs processing branch */
jne L(SPECIAL_VALUES_BRANCH)
# LOE rbx rbp r12 r13 r14 r15 edx xmm0 xmm7
/* Restore registers
* and exit the function
*/
/* Restore registers
* and exit the function
*/
L(EXIT):
movaps %xmm7, %xmm0
addq $72, %rsp
cfi_def_cfa_offset(8)
ret
cfi_def_cfa_offset(80)
movaps %xmm7, %xmm0
addq $72, %rsp
cfi_def_cfa_offset(8)
ret
cfi_def_cfa_offset(80)
/* Branch to process
* special inputs
*/
/* Branch to process
* special inputs
*/
L(SPECIAL_VALUES_BRANCH):
movups %xmm0, 32(%rsp)
movups %xmm7, 48(%rsp)
# LOE rbx rbp r12 r13 r14 r15 edx
movups %xmm0, 32(%rsp)
movups %xmm7, 48(%rsp)
# LOE rbx rbp r12 r13 r14 r15 edx
xorl %eax, %eax
movq %r12, 16(%rsp)
cfi_offset(12, -64)
movl %eax, %r12d
movq %r13, 8(%rsp)
cfi_offset(13, -72)
movl %edx, %r13d
movq %r14, (%rsp)
cfi_offset(14, -80)
# LOE rbx rbp r15 r12d r13d
xorl %eax, %eax
movq %r12, 16(%rsp)
cfi_offset(12, -64)
movl %eax, %r12d
movq %r13, 8(%rsp)
cfi_offset(13, -72)
movl %edx, %r13d
movq %r14, (%rsp)
cfi_offset(14, -80)
# LOE rbx rbp r15 r12d r13d
/* Range mask
* bits check
*/
/* Range mask
* bits check
*/
L(RANGEMASK_CHECK):
btl %r12d, %r13d
btl %r12d, %r13d
/* Call scalar math function */
jc L(SCALAR_MATH_CALL)
# LOE rbx rbp r15 r12d r13d
/* Call scalar math function */
jc L(SCALAR_MATH_CALL)
# LOE rbx rbp r15 r12d r13d
/* Special inputs
* processing loop
*/
/* Special inputs
* processing loop
*/
L(SPECIAL_VALUES_LOOP):
incl %r12d
cmpl $4, %r12d
incl %r12d
cmpl $4, %r12d
/* Check bits in range mask */
jl L(RANGEMASK_CHECK)
# LOE rbx rbp r15 r12d r13d
/* Check bits in range mask */
jl L(RANGEMASK_CHECK)
# LOE rbx rbp r15 r12d r13d
movq 16(%rsp), %r12
cfi_restore(12)
movq 8(%rsp), %r13
cfi_restore(13)
movq (%rsp), %r14
cfi_restore(14)
movups 48(%rsp), %xmm7
movq 16(%rsp), %r12
cfi_restore(12)
movq 8(%rsp), %r13
cfi_restore(13)
movq (%rsp), %r14
cfi_restore(14)
movups 48(%rsp), %xmm7
/* Go to exit */
jmp L(EXIT)
cfi_offset(12, -64)
cfi_offset(13, -72)
cfi_offset(14, -80)
# LOE rbx rbp r12 r13 r14 r15 xmm7
/* Go to exit */
jmp L(EXIT)
cfi_offset(12, -64)
cfi_offset(13, -72)
cfi_offset(14, -80)
# LOE rbx rbp r12 r13 r14 r15 xmm7
/* Scalar math fucntion call
* to process special input
*/
/* Scalar math fucntion call
* to process special input
*/
L(SCALAR_MATH_CALL):
movl %r12d, %r14d
movss 32(%rsp,%r14,4), %xmm0
call acosf@PLT
# LOE rbx rbp r14 r15 r12d r13d xmm0
movl %r12d, %r14d
movss 32(%rsp, %r14, 4), %xmm0
call acosf@PLT
# LOE rbx rbp r14 r15 r12d r13d xmm0
movss %xmm0, 48(%rsp,%r14,4)
movss %xmm0, 48(%rsp, %r14, 4)
/* Process special inputs in loop */
jmp L(SPECIAL_VALUES_LOOP)
# LOE rbx rbp r15 r12d r13d
/* Process special inputs in loop */
jmp L(SPECIAL_VALUES_LOOP)
# LOE rbx rbp r15 r12d r13d
END(_ZGVbN4v_acosf_sse4)
.section .rodata, "a"
.align 16
.section .rodata, "a"
.align 16
#ifdef __svml_sacos_data_internal_typedef
typedef unsigned int VUINT32;
typedef struct {
__declspec(align(16)) VUINT32 SgnBit[4][1];
__declspec(align(16)) VUINT32 OneHalf[4][1];
__declspec(align(16)) VUINT32 SmallNorm[4][1];
__declspec(align(16)) VUINT32 MOne[4][1];
__declspec(align(16)) VUINT32 Two[4][1];
__declspec(align(16)) VUINT32 sqrt_coeff[2][4][1];
__declspec(align(16)) VUINT32 poly_coeff[5][4][1];
__declspec(align(16)) VUINT32 Pi2H[4][1];
__declspec(align(16)) VUINT32 PiH[4][1];
__declspec(align(16)) VUINT32 SgnBit[4][1];
__declspec(align(16)) VUINT32 OneHalf[4][1];
__declspec(align(16)) VUINT32 SmallNorm[4][1];
__declspec(align(16)) VUINT32 MOne[4][1];
__declspec(align(16)) VUINT32 Two[4][1];
__declspec(align(16)) VUINT32 sqrt_coeff[2][4][1];
__declspec(align(16)) VUINT32 poly_coeff[5][4][1];
__declspec(align(16)) VUINT32 Pi2H[4][1];
__declspec(align(16)) VUINT32 PiH[4][1];
} __svml_sacos_data_internal;
#endif
__svml_sacos_data_internal:
/*== SgnBit ==*/
.long 0x80000000, 0x80000000, 0x80000000, 0x80000000
/*== OneHalf ==*/
.align 16
.long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000
/*== SmallNorm ==*/
.align 16
.long 0x2f800000, 0x2f800000, 0x2f800000, 0x2f800000
/*== MOne ==*/
.align 16
.long 0xbf800000, 0xbf800000, 0xbf800000, 0xbf800000
/*== Two ==*/
.align 16
.long 0x40000000, 0x40000000, 0x40000000, 0x40000000
/*== sqrt_coeff[2] ==*/
.align 16
.long 0xbdC00004, 0xbdC00004, 0xbdC00004, 0xbdC00004 /* sqrt_coeff2 */
.long 0x3e800001, 0x3e800001, 0x3e800001, 0x3e800001 /* sqrt_coeff1 */
/*== poly_coeff[5] ==*/
.align 16
.long 0x3d2EDC07, 0x3d2EDC07, 0x3d2EDC07, 0x3d2EDC07 /* poly_coeff5 */
.long 0x3CC32A6B, 0x3CC32A6B, 0x3CC32A6B, 0x3CC32A6B /* poly_coeff4 */
.long 0x3d3A9AB4, 0x3d3A9AB4, 0x3d3A9AB4, 0x3d3A9AB4 /* poly_coeff3 */
.long 0x3d997C12, 0x3d997C12, 0x3d997C12, 0x3d997C12 /* poly_coeff2 */
.long 0x3e2AAAFF, 0x3e2AAAFF, 0x3e2AAAFF, 0x3e2AAAFF /* poly_coeff1 */
/*== Pi2H ==*/
.align 16
.long 0x3fc90FDB, 0x3fc90FDB, 0x3fc90FDB, 0x3fc90FDB
/*== PiH ==*/
.align 16
.long 0x40490FDB, 0x40490FDB, 0x40490FDB, 0x40490FDB
.align 16
.type __svml_sacos_data_internal,@object
.size __svml_sacos_data_internal,.-__svml_sacos_data_internal
/* SgnBit */
.long 0x80000000, 0x80000000, 0x80000000, 0x80000000
/* OneHalf */
.align 16
.long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000
/* SmallNorm */
.align 16
.long 0x2f800000, 0x2f800000, 0x2f800000, 0x2f800000
/* MOne */
.align 16
.long 0xbf800000, 0xbf800000, 0xbf800000, 0xbf800000
/* Two */
.align 16
.long 0x40000000, 0x40000000, 0x40000000, 0x40000000
/* sqrt_coeff[2] */
.align 16
.long 0xbdC00004, 0xbdC00004, 0xbdC00004, 0xbdC00004 /* sqrt_coeff2 */
.long 0x3e800001, 0x3e800001, 0x3e800001, 0x3e800001 /* sqrt_coeff1 */
/* poly_coeff[5] */
.align 16
.long 0x3d2EDC07, 0x3d2EDC07, 0x3d2EDC07, 0x3d2EDC07 /* poly_coeff5 */
.long 0x3CC32A6B, 0x3CC32A6B, 0x3CC32A6B, 0x3CC32A6B /* poly_coeff4 */
.long 0x3d3A9AB4, 0x3d3A9AB4, 0x3d3A9AB4, 0x3d3A9AB4 /* poly_coeff3 */
.long 0x3d997C12, 0x3d997C12, 0x3d997C12, 0x3d997C12 /* poly_coeff2 */
.long 0x3e2AAAFF, 0x3e2AAAFF, 0x3e2AAAFF, 0x3e2AAAFF /* poly_coeff1 */
/* Pi2H */
.align 16
.long 0x3fc90FDB, 0x3fc90FDB, 0x3fc90FDB, 0x3fc90FDB
/* PiH */
.align 16
.long 0x40490FDB, 0x40490FDB, 0x40490FDB, 0x40490FDB
.align 16
.type __svml_sacos_data_internal, @object
.size __svml_sacos_data_internal, .-__svml_sacos_data_internal